[Scipy-tickets] [SciPy] #1469: arpack eigensolver returns incorrect eigenvector

SciPy Trac scipy-tickets@scipy....
Thu Jun 30 16:44:28 CDT 2011


#1469: arpack eigensolver returns incorrect eigenvector
----------------------------------+-----------------------------------------
  Reporter:  ElliotKaplan         |       Owner:  wnbell     
      Type:  defect               |      Status:  closed     
  Priority:  normal               |   Milestone:  Unscheduled
 Component:  scipy.sparse.linalg  |     Version:  0.9.0      
Resolution:  invalid              |    Keywords:             
----------------------------------+-----------------------------------------
Changes (by pv):

  * status:  new => closed
  * resolution:  => invalid


Comment:

 The results are correct:
 {{{
 print eval1[0], eval2[0], eval3[0]
 print abs(diff1.dot(evec1[:,0]) - eval1[0]*evec1[:,0]).max(), eval1[0]
 print abs(diff2.dot(evec2[:,0]) - eval2[0]*evec2[:,0]).max(), eval2[0]
 print abs(diff1.dot(evec3[:,0]) - eval3[0]*evec3[:,0]).max(), eval3[0]
 }}}
 prints
 {{{
 (-3.99903256458+0j) (3.99903256458+0j) (-0.000967435416024+0j)
 2.10942374679e-15
 2.83106871279e-15
 2.09534267107e-15
 }}}

 Arpack does not return the eigenvalues in the same order as `linalg.eig`
 nor all of them; the default is largest-magnitude only. See the
 documentation for details:

 http://docs.scipy.org/doc/scipy/reference/generated/scipy.sparse.linalg.eigs.html

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1469#comment:1>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list